icon-folder.gif   Conference Reports for NATAP  
  XVII International HIV Drug Resistance Workshop
June 10-14, 2008
Sitges, Spain
Back grey_arrow_rt.gif
NNRTI Mutation Rates With Failing EFV vs NVP in EuroSIDA
  XVII International HIV Drug Resistance Workshop
June 10-14, 2008, Sitges, Spain
Mark Mascolini
New nonnucleoside (NNRTI) mutations kept piling up as people continued taking failing efavirenz or nevirapine in a 154-person EuroSIDA analysis [1]. Mutations accumulated more slowly in people taking nevirapine than in those taking efavirenz, and mutations that confer resistance to the newest NNRTI, etravirine, cropped up more slowly than other NNRTI mutations. In a study of first-line efavirenz or nevirapine in the FIRST trial (reported separately by NATAP), significantly more NNRTI mutations arose when nevirapine failed than when efavirenz faltered [2]. But FIRST results on emergence of etravirine-specific mutations were equivocal.
Jens Lundgren and EuroSIDA colleagues scrutinized 326 genotype pairs from 154 people who had at least two genotypic resistance tests while taking nevirapine (174 genotype pairs) or efavirenz (152 genotype pairs). Eighty-seven people (57%) contributed one pair of genotypes, while the others contributed three or more genotypes. The first test had to reveal at least 1 NNRTI mutation. The EuroSIDA team calculated the number of new etravirine-specific mutations according to the IAS-USA list and figured the drop in etravirine activity according to the Rega resistance calculator.
At the time of the first genotypic test, 56% of EuroSIDA members were taking lamivudine (3TC), 47% stavudine (d4T), 42% didanosine (ddI), and 61% a protease inhibitor (PI). Combined use of a PI and an NNRTI suggests some of these people were taking multidrug salvage regimens. Median viral load at this point measured 4.1 log (a little over 10,000 copies), median CD4 count 224, and median number of NNRTI mutations 2 (range 1 to 9). Median interval between the first and second genotypes was 5 months. Average viral load did not change significantly in that time (+0.5 log, P = 0.09), and the group's average CD4 count rose by only about 6 cells during this interval. Viral load and CD4 changes were similar for people not taking a PI at any viral load measure.
Through 180 person-years of follow-up, 78 NNRTI mutations emerged between the first and second genotypes for an overall rate of 0.43 per year (95% confidence interval [CI] 0.36 to 0.51). This rate doubled an earlier estimate of zidovudine-induced mutation accrual in the EuroSIDA cohort. The new mutation rate was highest in people with virus predicted to be susceptible to the NNRTI being used at the first genotype (2.71 mutations per year), lower in virus predicted to have intermediate resistance at the first genotype (1.52 per year), and lowest in virus predicted to be resistant at the first genotype (0.36 per year).
Multivariate analysis figured almost a 4-fold faster buildup of NNRTI mutations per year (adjusted risk ratio [ARR] 3.77) with intermediate-resistant virus than with already-resistant virus (P = 0.002) and more than a 7-fold faster buildup per year (ARR 7.48) with susceptible virus than with already-resistant virus (P = 0.0001).
Nevirapine failure lowered the risk of NNRTI mutation accretion 38% compared with efavirenz failure, but that difference fell short of statistical significance (ARR 0.62, 95% CI 0.35 to 119, P = 0.10). When statisticians limited the analysis to people with one to three NNRTI mutations on the first genotype, nevirapine failure lowered the risk of NNRTI accumulation 58% compared with efavirenz failure, and that difference reached statistical significance (ARR 0.42, 95% CI 0.22 to 0.77, P = 0.006).
Forty-one new etravirine-specific mutations emerged during follow-up for an estimated rate of 0.23 per year (95% CI 17 to 30). Evolution of those mutations translated into an 8% per year absolute average reduction in susceptibility to etravirine, a seemingly moderate but highly significant change (P = 0.0001). Etravirine-specific mutations built up 80% more slowly in the 24 women analyzed than in the 130 men (P = 0.05), whereas etravirine mutations collected 3 times faster in people with sensitive or intermediate-resistant virus than in those with already-resistant virus.
Lundgren and colleagues suggested that "the impact of resistance accumulation over a long period on the response to etravirine-containing regimens remains to be seen." (from Jules: it is recommended not to leave patients failing on a regimen, or a NNRTI regimen, because they will accumulate mutations. Unless there is a good reason to remain on a failing regimen such as there are no other good treatment options to switch to. In particular if etravirine is available for use by a patient on a failing NNRTI regimen the patient should stop the NNRTI quickly so they do not accumulate mutations, so they can improve the response to etravirine.).
1. Cozzi-Lepri A, Clotet B, Paredes R, et al. The rate of accumulation of NNRTI resistance in patients kept on a virologically failing regimen containing NNRTI: a EuroSIDA study. XVII International HIV Drug Resistance Workshop. June 10-14, 2008, Sitges, Spain. Abstract 128.
2. MacArthur RD, Upller Hullsiek K, Peng G, et al. Failing therapy with efavirenz results in significantly fewer mutations limiting to etravirine than failing therapy with nevirapine: on-treatment analyses from the CPCRA FIRST study. XVII International HIV Drug Resistance Workshop. June 10-14, 2008, Sitges, Spain. Abstract 129.